TL;DR. Churn rate is the percentage of customers or recurring revenue you lose in a period; calculated as customers lost ÷ starting customers × 100. Healthy B2B SaaS keeps monthly churn below 1%; SMB ranges 3–7% monthly.
What churn rate actually measures in SaaS
Voluntary vs involuntary churn: two different problems
Churn does not come from a single source, and treating it as one metric drives the wrong interventions. Voluntary churn is the customer's active decision to cancel — and it usually signals product fit, pricing, onboarding failure, or competitive displacement. Involuntary churn is payment failure — expired card, bank decline, insufficient funds, antifraud block — and it represents 20–40% of total SaaS churn across the industry (Baremetrics Dunning Report 2025). These two require completely different tactics: dunning emails and card updater recover involuntary churn; engagement signals, NPS-triggered conversations, and CSM coverage address voluntary.
An operationally mature SaaS team reports both separately. A founder who only sees the headline churn number and increases the sales budget in response to what is actually a dunning problem will spend $200K on acquisition while leaking $80K monthly out the bottom of the bucket that a $5K dunning integration would have stopped.
Cohort analysis: the only honest way to see churn
Aggregate monthly churn is a lagging, averaged metric that hides the shape of the problem. The cohort view — retention by month of acquisition — shows three things aggregate churn cannot:
- Whether new cohorts churn faster or slower than older ones. If the D30 and D90 retention of Q1 2025 cohorts is lower than Q1 2024, product-market fit or onboarding is degrading.
- Which acquisition channel produces durable cohorts. An outbound cohort that closes at 12% monthly churn and a PLG cohort at 2% monthly churn both contribute to a blended 5% — obscuring that one channel is poison and one is gold.
- Where in the lifecycle churn concentrates. Month 1–3 churn is an onboarding problem. Month 6–12 churn is a value realization or renewal problem. Month 18+ churn is a competitive or pricing problem. Each has a different solution.
Onboarding as the primary churn lever
Across B2B SaaS, 30–50% of cancellations trace to customers who never achieved their first core value moment — they signed up, did not understand the product, and quietly churned within 90 days. The activation rate (customers who complete a defined milestone in their first 14 days) is the leading indicator of 3-month retention and is the single highest-leverage metric below LTV.
Benchmarks by Appcues (2025): products with activation rate >50% retain 70%+ of customers at 90 days; products below 20% activation retain less than 35%. A one-point improvement in activation typically produces a 0.3–0.6 point reduction in monthly churn. For a SaaS at $500K MRR and 3% monthly churn, that means $9K–$18K of monthly MRR preserved per activation-rate point gained — a better ROI than almost any acquisition channel.
Pricing and plan structure as churn modulators
Monthly billing inflates churn relative to annual billing by 3–5x in most SaaS categories. The reason is simple: monthly billing creates 12 cancellation moments per year; annual billing creates one. Customers who commit annually have also signaled deeper intent and are 40–60% less likely to churn in year 1 (ProfitWell 2024).
Tactics to migrate toward annual:
- Offer 15–20% discount for annual prepay (the effective discount is real; the involuntary churn savings alone justify 10–15%).
- Offer monthly billing only for the trial-to-paid conversion, then auto-suggest annual at month 3 when the customer has experienced value.
- For SMB motion, a 12-month commitment with a monthly-payment option (annual contract, monthly installments) provides the revenue protection of annual without the cash-flow objection.
Expansion MRR as the structural offset to churn
The reason NRR matters more than gross churn at scale is that expansion from existing customers can render churn economically irrelevant. A company with 3% monthly gross logo churn but 120% NRR is still growing its revenue base despite losing 3% of customers — because those it retains upgrade and expand faster than those it loses.
Expansion levers in SaaS:
- Seat-based expansion: more users at the same per-seat price.
- Usage-based expansion: more API calls, data volume, transactions, records — customers grow into higher tiers naturally.
- Upsell to higher tier: add premium features at a fixed additional cost.
- Cross-sell: adjacent products or modules that increase account ACV without acquiring a new customer.
Bessemer Cloud Index 2026 median NRR for top-quartile public SaaS: 124%. At that level, even 5% annual gross churn leaves the retained base growing. Below 100% NRR every cohort is shrinking — you are running backward.
CSM coverage model: who gets a human and who gets automation
Customer success coverage cannot be uniform across the customer base — economics prohibit it. The standard model tiers customers by ACV:
- High-touch: enterprise and strategic accounts (ACV >$20K–$50K). Dedicated CSM, regular QBRs, executive sponsorship, health score reviewed weekly. Churn here is existential — one $100K logo churning is 25 SMBs walking out.
- Tech-touch: mid-market (ACV $5K–$20K). Shared CSM with proactive playbooks triggered by health score dips, usage signals (logins, feature adoption) and NPS detractors. Email-first with escalation path.
- Digital/self-serve: SMB and PLG (ACV <$5K). Fully automated: product analytics triggers in-app nudges, email sequences, and community routing. CSM contact only on specific signals (usage drop >40% for 14 days).
A common scaling mistake: expanding the CSM team linearly with customer count, destroying unit economics. The right ratio: one CSM can manage $1.5M–$2.5M in ARR for high-touch, $4M–$8M for tech-touch, unlimited for digital via automation.
Churn rate is the percentage of customers (or recurring revenue) you lose in a period. In SaaS it is the metric that separates companies growing by accumulation from those bleeding out the bottom faster than they add at the top. A business adding 100 accounts monthly with 5% monthly churn on a base of 2,000 is pedaling in place: net growth stalls at 2,000 × (1 / 0.05) = a hard ceiling. The formula looks simple; its consequences are not.
Customer churn (logo churn) = (Customers lost in the period / Customers at the start of the period) × 100. If you started January with 500 customers and 15 canceled, your logo churn is 3.0%.
Revenue churn (MRR churn) = (MRR lost from cancellations + MRR lost from downgrades) / Starting MRR × 100. It almost always differs from logo churn: if large customers retain better than small ones, revenue churn is lower than logo churn. If the whales leave, it is higher. If you must report only one, report revenue churn: it is the number that ties back to the P&L.
Gross vs Net Retention (GRR vs NRR) — the metric VCs demand
Since 2023 the Bessemer Cloud Index and every Series A+ term sheet anchor on net revenue retention, not on logo churn.
GRR (Gross Revenue Retention) = (Starting MRR − Churn − Contraction) / Starting MRR × 100. GRR caps at 100%. It measures how much of the existing book you are preserving. Healthy B2B SaaS: 90%+; best-in-class: 95%+.
NRR (Net Revenue Retention) = (Starting MRR − Churn − Contraction + Expansion) / Starting MRR × 100. It can exceed 100% when expansion inside your base outpaces churn. That is negative churn. The public median is around 114%; vertical leaders exceed 120%. A company with 120% NRR doubles its base every 3.8 years without adding a single new logo.
2026 benchmarks by segment
Benchmarks only make sense against your ACV and motion:
- Enterprise SaaS (ACV > $50K, annual contract): 0.5–1.0% monthly, 6–12% annually.
- Mid-market (ACV $10K–$50K): 1.0–2.0% monthly, 11–22% annually.
- SMB (ACV < $10K, monthly plan): 3.0–7.0% monthly, 31–58% annually.
- B2C / prosumer: 5.0–9.0% monthly, 46–68% annually.
Vertical matters. Dev tools, cybersecurity and fintech retain better than marketing-tech, edtech and creator tools: switching cost and data lock-in are higher in the former.
Annualizing monthly churn — the 60% surprise
Many founders annualize "5% monthly" as 60% annual. The compound calculation yields 1 − (1 − 0.05)^12 = 46%. Conversely, 20% annual does not equal 1.67% monthly but 1 − (1 − 0.20)^(1/12) = 1.84%. If you project MRR at 24 months without the geometric form, your cohort drifts 10–20%.
From monthly to annual churn — conversion table
Compound formula: annual churn = 1 − (1 − monthly churn)^12. Typical values that appear in Google's People Also Ask:
| Monthly churn | Annual churn | Interpretation |
|---|---|---|
| 1% | 11.4% | Healthy B2B enterprise benchmark (Bessemer, ChurnZero). |
| 2% | 21.5% | Acceptable mid-market with inbound motion. |
| 3% | 30.6% | High threshold for SMB; review onboarding and activation. |
| 5% | 46.0% | Critical zone: you lose nearly half your base per year. |
| 7% | 58.2% | Prosumer/B2C without lock-in; plug the bucket before growing. |
| 10% | 71.8% | Freemium/creator tools with massive abandonment; no habit-forming product. |
References: Wall Street Prep, SaaS Churn Rate: Formula + Examples (2024); ChurnZero, Churnopedia: Monthly vs Annual Churn (2024); Redpoint Ventures / Tomasz Tunguz, The Cohort Retention Benchmark (reissued 2024).
Involuntary churn — the 30% almost no one reviews
Between 20% and 40% of total SaaS churn is involuntary: expired cards, failed payments, antifraud blocks. Recovery rates with dunning, card updater and smart retries range 38%–55%. If your gross monthly churn is 5% and 30% is involuntary, a decent dunning stack recovers 0.5–0.8 percentage points of monthly MRR. On $2M MRR that is $120K–$192K annually recovered without touching product or price.
Average customer lifetime — the LTV denominator
Average lifetime (months) = 1 / monthly churn. LTV (contribution margin) = ARPU × gross margin / monthly churn. At 3% monthly, average lifetime is 33 months. At 2%, it jumps to 50. That 1-point reduction expands LTV by 52%. That is why, at scale, investing in retention almost always dominates investing in acquisition: you compound the denominator.
How to use this simulator
Load your current MRR, monthly logo churn, monthly revenue churn and expansion MRR. The engine projects MRR at 12, 24 and 36 months under four scenarios: status quo, improved retention (−1 pp), expansion (NRR 115%+) and pricing (+10% ARPU). Each scenario returns the ending MRR and the LTV delta. Use it to prioritize: cutting 1 point of churn is usually worth 2–3× more than adding 1 point to acquisition.
What "good" looks like
Public best-in-class B2B SaaS: GRR ≥ 95%, NRR ≥ 120%, monthly logo churn ≤ 1%, involuntary ≤ 25% of total. If you meet three out of four you are compounding a business; if you meet none, acquisition will never catch up to churn.